IMPROVEMENT OF TAKAGI-SUGENO FUZZY MODEL FOR THE ESTIMATION OF NONLINEAR FUNCTIONS

Two new and efficient approaches are presented to improve the local and global estimation of the Takagi-Sugeno (T-S) fuzzy model. The main aim is to obtain high function approximation accuracy and fast convergence. The main problem is that the T-S identification method can not be applied when the membership functions are overlapped by pairs. The approaches developed here can be considered as generalized versions of T-S method with optimized performance. The first uses the minimum norm approach to search for an exact optimum solution at the expense of increasing complexity and computational cost. The second is a simple and less computational method, based on weighting of parameters. Illustrative examples are chosen to evaluate the potential, simplicity and remarkable performance of the proposed methods and the high accuracy obtained in comparison with the original T-S model.